Performance Analysis and Power Allocation of Multi-Pair Two-Way Massive MIMO Full-Duplex DF Relaying Systems Under Rician Fading Channels

Abstract

This paper investigates a two-way full-duplex decode-and-forward relaying system under Rician fading channels with imperfect channel state information. In this system, multiple pairs of users exchange information via a relay equipped with a large number of antennas. Both the maximum-ratio combining/maximum-ratio transmission (MRC/MRT) and the zero-forcing (ZF) processing methods are considered at the relay. We derive the closed-form approximations for the spectral efficiencies (SEs) of the considered system. Then, we study the power-scaling laws for two typical scenarios, which show that as the number of the relay antennas <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$M$</tex-math></inline-formula> grows large, the SEs converge to positive limits when the power scaling factors <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha _{\mathrm{u}}$</tex-math></inline-formula> of the users and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha _{\mathrm{r}}$</tex-math></inline-formula> of the relay satisfy <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha _{\mathrm{u}} = 1$</tex-math></inline-formula> or <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\alpha _{\mathrm{r}} = 1$</tex-math></inline-formula> , i.e., the transmit power of the users or the relay is scaled down proportionally to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$1/M$</tex-math></inline-formula> . Furthermore, for MRC/MRT and ZF processing methods, the optimal power allocation scheme we proposed largely improves the sum SEs compared with the average power scheme. Simulation results verify the correctness of the derived results.